GB2038297A - Producing sulphur-nitrogen groups - Google Patents

Abstract

Sulphur nitrogen groups are made by reducing species SaNbXc where X is monovalent group or <1>/n of n-valent group, a>/=b, a>/=c and a is up to 6. In particular compounds such as S4N4, S5N5<+> salts and poly(sulphur nitride), are made by reducing S4N3Cl or related compounds with an ions (e.g. iodide) or metals (e.g. silver). Poly(sulphur nitride) may be made by passing S4N3Cl vapour or S5N5FeCl4 vapour through sulphidised silver wool and condensing the vapour.

Description

SPECIFICATION Producing sulphur-nitrogen groups This invention relates to the production of sulphur-nitrogen (SN) groups, and to compounds containing them, such as S4N4, Sun5+ salts and poly (sulphur nitride) compounds and to methods of preparing them.

The poly(sulphyr nitride) compounds include derivatives containing halogen which are electrically conducting polymers with no metal atoms, and can be compressed rather like graphite to form black semiconducting blocks. In some cases, the electrical resistance increases with voltage to a maximum and then falls with further increasing voltage.

The invention consists in producing SN groups (either in the form of S4N4 molecules or instantaneously reacting to form a compound containing an -S-N- grouping), by reducing the species SaNbXC is a monovalent group such as nitrate or a halogen, preferably Cl, (or 11n of an n-valent group), aMb and a3c (and preferably bsc) and a is up to 6, e.g. NSX, S3N3X3, S4N3+, S5N5+ S3N3X or S6N4X2. This reduction may be done in the presence of a reagent which will react with any S4N4 produced to form a further compound containing an -S-N grouping.The reducing agent may be for instance a metal (such as iron (e.g. filings) or silver (e.g. wool) or titanium (e.g. sponge)) or may contain reducing anions such as chalcogenide, iodide, oxalate, hypophosphite, thisulphate or metabisulphite, and may be dissolved or suspended in a non-aqueous medium, such as nitromethane CH3NO2, dichloromethane CH2CI2, or methanol CH3OH, or may be used in certain instances in the vapour phase. Aqueous solutions (e.g. concentrated ice-cold aqueous Na2S205 with S3N2CI or S4N3CI) may be advantageous in certain circumstances. In some cases, reaction may occur (at least in part) even in the absence of solvent.

Examples of the species SaNbCIC are S4N3CI, S3N2CI (which are easy to make and have gratifying low sensitivity to moisture), and S3N2C12 and S3N3CI3. The preferred compounds are S4N3CI and S3N2CI, but S3N2C12 can be used in this preferred mode by converting it first to S3N2CI, for example by heating to about 80"C in vacuo, when NSCI and SCI2 are also formed, or by treatment with formic acid HCOOH.S3N3CI3 for this purpose includes NSCI (which is normally in equilibrium with S3N3CI3), and also the ion CIS-N=SCI+ (formed for example by the reaction /3S3N3CI3+SCI2+AICI3 w [N (SCI)2] +AICl4-.) Examples of compounds containing an -S-N-grouping, which may be produced according to the invention, include (apart, of course, from S4N4 itself) 1 ,2,5-thiadiazoles; other heterocyclic and chain compounds such as R1 (SN)nR2 where R1 and R2 are chain-terminating groups e.g. (i) R3(SN)2SR4 and (ii) R2C=N(SN)2=CR2, which are highly coloured compounds whose hue varies with R3-6; and (iii) polymers (-SN-)X and (SN Halogen,), where c < 1 and xis large; these polymers contain sulphur-nitrogen chains and are notable for an electrical conductivity approaching that of mercury, strongly anisotropic metallic properties and super conductivity below 0.26K. For some of the above products and under the right conditions (e.g. sufficiently high temperature) the compound S4N4 reacts rapidly so that even if S4N4 itself were produced according to the invention, it would be practically instantaneously reacted to form the desired compound, so that practically no free S4N4, which is potentially explosive, is present at any time.

Where the desired compound in fact is S4N4, this may be purified by recrystallisation (e.g. from dichloromethane, 1 ,4-dioxan, benzene ortoluene).

The invention in a further aspect provides the new compound(s) (S3NblC)x where b is from 1.9 to 2.3 (preferably z.1 to 2.3), c is from 0.4 to 1.3 (preferably 0.4 to 0.6), and x is large. At the higher iodine contents, free iodine may be present.

The invention also provides a method of preparing the compound (S3 NblC)x, comprising contacting S4N3CI vapour with an iodide and cooling the contacted vapour to below room temperature, preferably 0-15"C. The iodide may be in the form of a salt, preferably an alkali metal salt, and may be held at at least 1 500C, preferably at least 200"C, e.g. around 250"C.

In a further aspect, this invention relates to poly(sulphur nitride) compounds. Such compounds, which may be doped with a proportion of halogen, are electrically conducting polymers. Some of these compounds become superconducting at low temperatures. They may be useful as the contact material in Schottky diodes on semiconducting devices, for example as the contact material on GaAs solar cells.

According to the present invention, poly (sulphur nitride) is made by contacting S4N3CI) vapour with silver sulphide and allowing the contacted vapour to condense (at -196 C) to give S2N2 which polymerises at room temperature to give (SN)X. The silver sulphide is preferably held at at least 2500C. The silver sulphide may be derived from metallic silver which is sulphided by the contact with the S4N3CI vapour.

Also according to the present invention, poly (sulphur nitride) is made by heating S5N5FeC4 in an inert atmosphere to at least 10000 and condensing the resulting vapour at between -100 C and 20"C, preferably from 0 C to 15"C. Preferably, in order to lessen the proportion of S4N4 which is formed and which condenses at a slightly different temperature, the vapour is contacted with silver before being condensed. By this method, thin layers of poly (sulphur nitride) can be obtained, useful for a contact material in semiconducting devices.

The invention will now be described by way of example. The accompanying drawing illustrates apparatus used in Example 19.

Example 1 In Examples 1 to 7, S4N4 is produced from S3N2Ci.

S3N2CI (0.10g, 0.63 mmole) was ground with 0.129 (0.75 mmole) of dry sodium iodide Nal in a glove-box, at a temperature of about 20"C. After 15 minutes, examination of the infra-red absorption peaks of the reactants showed that no S3N2CI remained, and that S4N4 was present. Thus, the following reaction scheme is indicated where S3N21 is a possible reaction intermediate: 8S3N2CI + 8NaCI + 8S3N21 8S3N2194S4N4 + Ss + 412 A soxhlet extraction of the products, using 1,4-dioxan yielded S4N4, as well as iodine and some hydrolysis products.

Example 2 A suspension comprised S3N2CI (0.30g, 1.9 mmole) in 10ml of nitromethane CH3NO2. (In another experiment, the volume of (refluxing) nitromethane was a third of this, with some diminution in yield.) Into the suspension was stirred Na2S2O5 (0.55g, 2.9 mmole) under nitrogen at 90 C. A reaction occurred, and appeared to be complete after about 15 minutes. After filtration, washing the product with ice-cold water and recrystallisation from dichloromethane (to remove sulphur), some 0.12g of S4N4 (identified by infra-red analysis and nitrogen analysis) were obtained, representing a yield of 69%.

Example 3 S3N2CI was treated with an aqueous solution of sodium iodide. Liberated iodine was extracted using tetrachloromethane, and infra-red analysis of the products showed that S4N4, as well as some hydrolysis products, had formed.

Example 4 Impure S3N2CI (2.759, 17.2 mmole obtained by heating 5.3g S3N2CI2, 27.2 mmole) was stirred (2 hours, 100"C) with iron turnings (0.96g, 17 mmole) suspended in nitromethane (50ml). After filtration while hot and evaporation of solvent under reduced pressure (to 15ml),the product (1.1 g) was recrystallised from dichloromethane to give S4N4 (0.79,28% yield from S3N2CI2).

Example 5 Example 4 was repeated using 1 .67g, 10.5 mmole of the S3N2CI and 0.6g of the iron turnings (10.7 mmole) for 2 hours in 30ml refluxing nitromethane. The mixture was cooled and filtered, and the resultant brown solid extracted from 30 ml tetrachloromethane to give 0.259 tetrasulphur tetranitride S4N4 is a yield of 26%.

Example 6 S3N2CI (5.0g, 31.3 mmole) was reacted with sodium iodide (5.49, 36.0 mmole) in 20ml methanol at 0 C for 1/2 hour. Elemental iodine was quantitatively removed using aqueous sodium thiosulphate, leaving a black mixture of S4N4 and iodinated poly (sulphur nitride), the formation of the latter being favoured by the presence of water. From this mixture S4N4 was isolated in a yield of about 40% by extraction using dioxan as solvent. However, the S4N4 was contaminated with sulphur S5 as shown by thin-layer chromatography, using which technique (employing anhydroys silica gal or alumina with hexane as elutant) good recovery of S4N4 is possible.Excessive moisture can hydrolyse the iodinated polymer.

Example 7 Example 6 was repeated, using 6.0g, 37.6 mmole of the S3N2Cl and 8.45g of the anhydrous sodium iodide (56.4 mmole) stirred for 1 hours in 30ml dry methanol at 0,. Anhydrous sodium thiosulphate Na2S203 (8.9g, 56.4 mmole) in 30ml dry methanol was added to remove liberated iodine, and the mixture was stirred for 18 hours. The filtered yellow-green product was extracted with 1 ,4-dioxan to give 1 .8g of a mixture which analysed as 66% tetrasulphurtetranitride S4N4 and 34% sulphur by weight.

Example 8 In Examples 8 and 9, S4N4 is produced from S3N2CI2, chlorocyclotrithiadiazenium chloride. In Example 9, a further product containing sulphur-nitrogen groups also arises.

S3N2CI2 was refluxed for 8 hours in a nitromethane solution with Na2S205. The products were washed with water. Infra-red analysis indicated S4N4 together with hydrolysis products.

Example 9 A mixture of chlorncyclotrithiadiazenium chloride (S3N2CI2) (7.0g, 35.9 mmole) and iron turnings (2.59, 44.6 mmole) was stirred for 6 hours in refluxing nitromethane (40 ml). Intermediate green and yellow colours indicated that the reaction probably proceeded via cyclotrithiadiazenium chloride (S3N2CI) and then cyclotetrathiatriazenium chloride (S4N3CI) and/or cyclotetrathiatriazenium tetrachloroferrate S4N3+FeCl4.

After cooling and evaporation of the nitromethane under reduced pressure, the resultant solid was extracted in tetrachloromethane (50 ml) to give tetrasulphur tetranitride (0.6g, 18% yield) and a residue of cyclotetrathiatrizenium tetrachloroferrate (III) which was the main product.

Example 10 In Examples 10 and 11, S4N4 is formed from S3N3Cl3.

S3N3C13 (0.20g, 0.82 mmole) was dissolved in dry 1,2 dimethoxyethane CH3-O-CH2-CH2-O-CH3 (10 ml). This solution was pale green. Then iron (0.05g, 0.89 mmole) was added, in the form of commercial activated sponge, at room temperature. The solution deepened in colour, becoming deep orange-red after 5 - 10 minutes. It was verified that the solution contained S4N4 by thin-layer chromatography of a sample of the solution in parallel with a solution of known S4N4. The chromatography was performed on a silica gel plate, using carbon disulphide CS2 as eluant and detecting the S4N4 by ultraviolet light.

Example 11 Example 10 was repeated but this time using S3N3C13 (7.09, 28.6 mmole) and iron (2.59, 44.8 mmole) in the form of turnings. The mixture was stirred for 1 hour in 1,2 dimethoxyethane (40 ml). An exothermic reaction occurred and thin layer chromatography showed S4N4,S4N2 (and probably S4N3+) to be present. The mixture was refluxed for 1 hour, cooled and the dimethoxymethane evaporated under reduced pressure.

Tetrasulphurtetranitride was extracted from the crude product using 50 ml tetrachloromethane (50 ml) and recrystallised from benzene (65 ml) to give tetrasulphur tetranitride (2.65g, 67% yield).

Example 12 In Examples 12 to 14, S4N4 is produced from S4N3CI (cyclotetrathiatriazenium chloride).

S4N3CI was stirred with Na2S205 in molar ratio 1:1:15 in refluxing nitromethane for 1 hour. The products were washed with water. Infra-red analysis indicated S4N4togetherwith hydrolysis products. Recrystallisa tion from dichloromethane gave S4N4 (25% yield).

Example 13 A mixture of S4N3CI (5.0g, 24.3 mmole) and iron turnings (1.5g, 26.8 mmole) was stirred for 11/2 hours in refluxing nitromethane (100 ml) to give a clear red solution. This was filtered hot (grade 1 sinter). Crude S4N4 crystallised out on cooling. The filtered product was extracted with 50 ml tetrachloromethane and recrystallised from benzene (50 ml to give 1.869, S4N4 in 55% yield.

Example 14 A mixture of S4N3CI (1.0g, 4.9 mmole) and sodium thiocyanate (0.5g, 6.2 mmole) was stirred for 1 hour in refluxing CH3O(CH2)2OCH3 (30 ml). After cooling, the orange product was filtered and extracted from 30 ml tetrachloromethane to give S4N4 (0.229,33% yield).

Example 15 Preparation of S4N4-norbonylene derivative.

S4N3CI (4.0g, 19.5 mmole) was stirred for two hours with iron turnings (1.0g, 17.9 mmole) suspended in refluxing nitromethane (40 ml). The mixture was cooled, filtered and the solid Soxhlet extracted with dioxan (150 ml.) Norbonylene (7.09, 74.5 mmole). was added and the mixture stirred for eighteen hours at 50"C.

Crude S4N4(C7H10)2 (219) came out of solution, was filtered and recrystallised from chloroform:ether (25 ml: 5 ml) to give S4N3(C7H10)2 (1 .4g, 26% yield based on S4N3CI):

U.K. Patent No. 1445968 has claims which are relevant to Example 15.

Example 16 In this Example, we proceed direct to the formation of a Lewis acid adduct without intermediately isolating S4N4. The adduct in question is (S4N4)2SnCl4, i.e. the sulphur nitride/tin tetrachloride adduct. The adduct can be reduced to tetrasulphu r tetramide.

A mixture of S4N3CI (2.2g, 10.7 mmole) and iron turnings (0.70g, 12.5 mmole), suspended in nitromethane (100 ml), was heated under reflux for 90 minutes to give a clear red solution. After filtration through a grade 1 sinter filter and cooling, tin tetrachloride (0.9 ml) was added. A crimson precipitate formed at once. It was the desired adduct. The mixture was stirred overnight at room temperature to complete the reaction and filtered.

The adduct was washed with dilute HCI (20 ml), ethanol (20 ml) and diethyl ether (20 ml) and characterised by analyses and its infra-red spectrum. The yield was 1.6g, 63% based on the equation: 4S4N3CI + 2Fe 1/2S8 1/2S8 + 3S4N4 + 2FeCI2 2S4N4 + SnC14 ) (S4N4)2SnCI4 It is also possible to use 'in situ' S4N4 for other syntheses, such as of S4N4H4 or S4N4/norbornylene adduct, but sulphur contamination can be a problem in these cases. In addition, the adduct (S4N4)2SnCi4 suspended in benzene can be reduced, by SnCI2-2H2O, to S4N4H4.

Example 17 In this Example we prepare cyclopentathiapentazenium tetrachloroferrate (III) S5N5tFeCI-4- A mixture of trichlorocyclotrithiatriazene (NSCl)3 (6.0g, 24.5 mmole) and iron turnings (2.5g, 44.6 mmole) was stirred for 18 hours in nitromethane (25 ml). Liquid SO2 will also work similarly. Thin layer chromatography of the solution during the early stages of the reaction showed the presence of S4N4. The nitromethane was evaporated to low bulk, and the yellow product filtered and extracted with 50 ml thionyl chJoride (50 ml). Recrystallisation from furtherthionyl chloride gave S5N5+FeCI4- (2.62gm 42%). Use of aluminium instead of iron turnings gives S5N5+AICI4-.Attempts to replace FeCI4- by other anions (e.g.

NCS-, PF6 and C6H5O-) by metathesis have been shown, by thin-layer chromatography, to result in decomposition with formation of S4N4.

Example 18 in Example 18, a black conducting iodinated sulphur nitride polymer is produced in the vapour phase from S4N3CI and Nal. A glass pyrolysis tube (60 cm x 3 cm external diameter) was connected via a tap, ball and socket joint and liquid nitrogen trap to a vacuum line fitted with a mercury diffusion pump. Cyclotetrathiatrizenium chloride S4N3CI (1.0g, 4.8 mmole) was contained in a pre-weighed glass bucket at the bottom of the pyrolysis tube. An oil bath warmed the bottom part of the tube. Sodium iodide supported on glass wool (details below) was packed loosely (to a height of 20 cm) in the tube above the S4N3CI.The plug of glass wool/sodium iodide was surrounded by a heating mantle consisting of 117 loops at 2 mm spacing of nichrome wire of resistance 26.25Q/m cemented helically inside a glass tube (4.5 cm inside diameter) sliding easily over the pyrolysis tube. This design provides a good heat profile (i 1"C over 4 14cm at the centre of the mantle) and a temperature of 250 (+ 5)'C over the whole tube at an applied voltage of about 145V. The glass wool (6.09g; fibre diameter 3 - 5ym) was vacuum dried (300"C, 12 hours) and shaken with a solution of vacuum dried (300"C, 12 hours) sodium iodide (ca. 2g) in dry methanol (35 ml).After evaporation of the methanol, the glass wool was vacuum dried (20"C, 4 hours) and weighed (7.359).

The top of the pyrolysis tube was surrounded by a glass cooling jacket (14cm x 6cm external diameter) with circulating alcohol (10"C). The system was gradually (over 15 minutes) evacuated to below 5 x 10-3 torr using rotary vacuum pump assisted by the mercury diffusion pump. The mantle was heated to 250"C and then the oil bath gently heated to a maximum of 130"C. Pumping was continued for 3 - 4 hours, until the cyclotetrathiatriazenium chloride had mostly vaporised.

During the reaction a compact layer of black, flaky material condensed in the cold zone. This material, in a yield of 0.9 - 1 .04g, was removed in a dry, inert atmosphere to avoid surface hydrolysis.

Atypical analysis gave S, 49.90; N, 15.59; 1,34.51; which corresponds to S3N21510.53. The infra-red spectrum and X-ray data are shown in Tables 1 and 2. The d.c. resistance of a plug (3.1 mm diameter x 3.1 mm) of the material (compressed to 2 gigapascals) at room temperature between Pt electrodes was found to be about 10Q. The black polymeric product had a compact and brittle constitution and could be scraped off the pyrolysis tube wall as rigid flakes. It had a semi-metallic lustre which tarnished only slightly over several months when kept in a sealed sample bottle. On heating above 40"C at below 10-6 mm the produce decomposes to S4N4, sulphur and iodine.The polymer was moisture sensitive and when left exposed to moist air over 3 - 4 days it hydrolysed fairly rapidly to give a greyish powder with release of iodine.

Examination of the product under the microscope (x 175 and x 500) showed that the slightly surface-hydrolysed material had a dull, pitted surface and was largely non-crystalline. Infra-red absorptions, Table 1, also revealed some hydrolysis (absorptions around 3200 and 1390 cm-l are due to VN-H) and that some decomposition to tetrasulphur tetrantride had occurred. Other peaks were within the typical range of S-N absorptions (1500 cm-1 to 600 cm-1) but differed from those in S4N4 and (SN)X. The unhydrolysed product under the microscope had a shiny bronze fibrous appearance.The proportion of iodine in the polymer varied with the temperature of condensation, being 1.3 atoms per 3 S atoms at -77 C, and 0.5 atoms per S atoms at +IOoC. The higher value is likely to include elemental iodine which would not condense (by contrast) at +10'C.

The mass spectrum is typical of the general breakdown of a sulphur-nitrogen ring or chain and is similar to that for (SN)x and S4N4.

Confirmation of the low degree of crystallinity in the polymer is afforded by the X-ray powder diagrams.

Diffraction lines were only faintly visible even after 48 hours' exposure to X-rays. The X-ray powder data, Table 2, did not show any crystalline iodine in the polymer. Iodine was probably present in the polymer as anions and/or termnal S-l. Hydrolysis of the S-I bonds would explain the formation of free iodine. Sulphur does not appear to be an impurity in the original polymer but there is evidence for its presence in the residue after heat treatment. The diffraction lines of the polymer were substantially different from (SN),c.

This method is rapid, single-stage and avoids isolation of potentially explosive tetrasulphur and disulphur dinitride.

TABLE 1 Infra-red absorption of the compounds made according to the foregoing example: 3200-3100w; 1390 m; 1190vw,br; 1110 m, br; 1030vw,br; 923 ms; 850 - 750 w, br; 705 ms, br; 610 m, br; 563 ms; 460 w, br; 348 ms.

br = broad; m = medium; s = strong; v = very; w = weak TABLE 2 X-ray diffraction data (d-spacing) of block poly(sulphur nitride) iodide. (Si calibration ASTM No. 5-0565.) 4.782; 4.575; 4.361; 3.921; 3.869; 3.282; 3.237; 3.163; 3.135; 3.063; 3.012, no further lines resolved.

Example 19 This Example demonstrates that cyclotetrathiatriazenium chloride S4N3CI vaporised at 130"C reacts with silver at 300"C to give AgCI, Ag2S and S2N2, the last of which may be isolated to polymerise to a chlorine-doped (SNIP (poly (sulphur nitride)).

Cyclotetrathiatriazenium chloride (S4N3CI) was prepared from S3N2CI2 and recrystallised from thionyl chloride. The thionyl chloride was purified by twofold distillation from triphenylphosphite. The fine yellow crystals were dissolved (about 1 gilt ml) in refluxing thionyl chloride and cooled slowly to room temperature, under dry nitrogen. Care was taken that the crystallisation temperature did not fall much below 20"C since, when cooled in a refrigerator (about -5 C) or evaporated under reduced pressure, the solution usually darkened and the precipitated yellow crystals turned dark green. Such contaminated product contained four extra infra-red absorptions at 965,945,710 and 590 cm-l.

Silver wool of 99.99% purity, about 50pm diameter, was coiled and formed into a plug. The plug was degreased with hexane in a Soxhlet extractor, washed in concentrated ammonia and dried in vacuo.

An all-glass pyrolysis apparatus (Figure 1) was set up consisting of a train of two U-tubes 6,7 and a pyrolysis tube 1 within transparent furnaces 4, 5. A jointless system with seal-off sections and seal-off constrictions b, c, d, e and f was chosen since it allowed a large proportion of the apparatus to be flamed out under vacuum, and thus to achieve a thorough outgassing. It also facilitated the transfer, isolation and storage of the volatile, moisture-sensitive products.

A charge 2 of S4N3CI (0.460g, 2.24 mmole) was placed in the pyrolysis tube 1, which was plugged with 1.3739 (12.73 mmole) of the silver wool 3. The inset diagram shows the temperature distribution along the axis of the pyrolysis tube 1.

The reaction progressed in three distinct stages (i) For about two hours after the temperature of the S4N3CI had steadied at 130"C it remained confined to the silver zone (formation of Ag2S and AgCI). (ii) In the second stage two simultaneous processes were observed; the formation of a blue transparent layer in zone Z1 and the formation of white S2N2 deposits in zone Z2 and zone Z3. After about three hours, when more than a half of the initial S4N3CI charge was spent, the appearance of a brown substance in zone Z3 heralded the third stage, which was characterised by continuing formation of S2N2 accompanied by simultaneous production o-, .. ,te,1s;-leiy coloured (orange to brown) by-products which contaminated the disulphur dinitride.Towards the end of the reaction an orange-yellow ring formed at the top of zone Z3 (later identified as (NSCI)B). (In a further experiment, aiready-sulphidised silver wool was used in place of the pure wool. Here, the first stage was absent and an immediate formation of the blue deposit (characteristic of the commencement of stage ii) occurred as soon as the temperature of the S4N3CI charged reached about 100"C. Partly or wholly replacing the silver sulphide, silver selenide and/or silvertelluride can be used, thus implanting a certain proportion of selenium/tellurium into the polymer.) The total yield of the S2N2 (about 55%) along with some other volatile constituents was collected in the U-tube 7 cooled at - 1 96'C and sealed off.The U-tube 6 was held at OOC. The S2N2 crystals grew at room temperature and were left to polymerise over a period of 6 weeks. After that time all the deposits, irrespective of their original colour, turned dark blue. The (SN)x crystals which grew partly on clean glass surfaces and partly on the dark blue felt-like showed brass yellow colour and a high metallic lustre. They range from sub-millimetre to millimetre sizes, and a large proportion hakd well developed facets and belonged decidedly to the 'less fibrous' class.

A well-developed (SN)x crystal was examined by electron probe micro-analysis simultaneously with another (SN)x crystal, prepared by the usual route from S4N4. The following results were obtained: Electron probe micro-analysis of (SN), < Crystals From S4N3CI From S4N4 Sulphur Chlorine Sulphur Chlorine wt% wt% wt% wt% One spot 69.65 0.03 69.36 0.03 Another spot 69.49 0.07 69.34 0.00 spot Yet another spot 69.47 0.00 69.32 0.00 69.54 0.05 69.34 0.01 Theory 69.59 Example 20 In Examples 20 and 21, poly(sulphur nitride), (SN)x, in the form of a thin layer is made from cyclopentathiapentazenium tetrachloroferrate (III).

Silver wool of 99.99% purity, about 50ym diameter, was coiled and formed into a plug. The plug was degreased with hexane in a Soxhlet extractor, washed in concentrated ammonia and dried in vacuo.

Cyclopentathiapentazenium tetrachloroferrate (III) was prepared by the reaction of (NSCI)3 with iron. The apparatus used consisted of an upright sublimation tube fitted with a water-cooled cold finger depending from the top of the tube. The sublimation tube was connected via all-glass joints and a liquid nitrogen trap to a vacuum line fitted with a mercury diffusion pump. A heating mantle, of nickel-chrome wire coiled longitudinally around a split heat-resisting glass tube, was fastened around the sublimation tube. The heating mantle was arranged to produce a temperature of 240"C around the sublimation tube. An oil bath warmed the bottom part of the tube.

Cyclopentathiapentazenium tetrachloroferrate (III) (0.27g, 0.63 mmole), prepared as set forth in Example 17, was placed in the bottom of the tube, between which and the cold finger was a plug (0.259, 23.2 mmole) of the silver wool. The system was kept evacuated to below 1 0-3 torr by the mercury diffusion pump. With the mantle at 240"C the oil bath was gently heated to a maximum of 130"C. The silvverwool rapidly blackened due to the formation of Ag2S and after 6 hours a blue film of (SN)x became apparent on the cold finger. After 11 hours, the (SN)x on the lower part of the cold finger and assumed a bronze sheen.Although after 16 hours the S5N5FeC14 charge had not been completely exhausted, there was a substantial bronze layer of polymer adhering to the lower part of the cold finger, with a blue film of polymer above. In this region there were a few small green patches (identified from their infra-red spectra as a S4N4/(SN)X mixture) presumably due to incomplete catalytic conversion by the Ag2S. The residue of unreacted S5N5FeCl4 had assumed a khaki colour in the course of the experiment and S4N3FeCI4was also present.

Another experiment using S5N5AICI4 gave no reaction under similar conditions.

Example 21 Example 20 was repeated using S5N5FeCI4 but omitting the silver wool. The result was a green layer of S4N4 and (SN)x on the lower part of the cold finger with a blue film of (SN)x above this. Thus some (SN)x can be produced from S5N5FeCI4 even when no silver wool as catalyst is employed. This is not the case with S4N4 and S4N3CI.

Thus, Examples 20 and 21 represent a new route to films of (SN)x, avoiding the use of S4N4 or S2N2.

Claims (27)

1. Producing sulphur-nitrogen groups by reducing the species SaNbXcwhereX is a monovalent group or 1/n of an n-valent group, a3b and a3c and a is up to 6.

2. Producing sulphur-nitrogen groups according to Claim 1, wherein X is nitrate or a halogen.

3. Producing sulphur-nitrogen groups according to Claim 2, wherein X is chlorine.

4. Producing sulphur-nitrogen groups according to any preceding claim, wherein b .

5. Producing sulphur-nitrogen groups to Claim 1, 2 or 3, wherein SaNbXC is any one of NSX, S3N3X3, S4N3+, S5N5X or S6N4X2.

6. Producing sulphur-nitrogen groups according to Claim 5, wherein SaNbXC is S4N3CI.

7. Producing sulphur-nitrogen groups according to Claim 5, wherein SaNbXC is S3N2Cl.

8. Producing sulphur-nitrogen groups according to any preceding claim, wherein the reducing agent is a metal or reducing anions.

9. Producing sulphur-nitrogen groups according to Claim 8, wherein the reducing agent is iron, silver, titanium, chalcogenide, iodide, oxalate, hypophosphite, thiosulphate or metabisulphite.

10. Producing sulphur-nitrogen groups according to any preceding claim, in the presence of a reagent which will react with any S4N4 produced to form a further compound containing an -S-N- group.

11. Producing sulphur-nitrogen groups according to any preceding claim, wherein the reducing agent is suspended or dissolved in a non-aqueous medium.

12. A compound of the formula (SaNblc)x where b is from 1.9 to 2.3, c is from 0.4 to 1.3 and is large.

13. A compound according to Claim 12, wherein b is from 2.1 to 2.3.

14. A compound according to Claim 12 or 13, wherein c is from 0.4 to 0.6.

15. A method of producing a compound according to Claim 12, 13 or 14, comprising contacting S4N3CI vapour with an iodide and cooling the contacted vapour to below room temperature.

16. A method according to Claim 15, wherein the iodide is held at at least 150"C.

17. A method according to Claim 16, wherein the iodide is held at at least 200 C.

18. A method according to Claim 15,16 or 17, wherein the iodide is in the form of a salt.

19. A method of producing poly(sulphur nitride) comprising contacting S4N3CI vapour with silver sulphide and allowing the contacted vapour to condense to give S2N2, which is then allowed to polymerise at room temperature.

20. A method according to Claim 19, wherein the silver sulphide is held at at least 250"C.

21. A method according to Claim 19 or 20, wherein the silver sulphide is derived from metallic silver which is sulphidised by contact with the S4N3CI vapour.

22. A method according to Claim 19 or 20, wherein the silver sulphide is partly or wholly replaced by silverselenide andior silver tel luride.

23. A method of producing poly(sulphur nitride), comprising heating S5N5FeCI4 in an inert atmosphere to at least 100"C and condensing the resulting vapour at between -100"C and 20"C.

24. A method according to Claim 23, wherein the resulting vapour is condensed at from 0'Cto 15"C.

25. A method according to Claim 23 or 24, wherein the vapour is contacted with silver before being condensed.

26. A method according to Claim 23, 24 or 25, wherein the condensation is performed on a cooled surface, whereby the poly(sulphur nitride) is produced in the form of a layer.

27. A method of producing sulphur-nitrogen groups according to Claim 1, substantially as hereinbefore described with reference to any one of Examples 1 to 21.